A novel nighttime light downscaling index approach for generating annual high-definition nighttime light products

Satellite-observed nighttime lights (NTL) are pivotal for urban studies; however, the coarse spatial resolution (∼500 m) of the VIIRS NTL limits their utility in capturing fine-gridded intra-urban dynamics. We proposed a Nighttime Light Spatial Downscaling Index (NTLDI) approach to generate 100 m resolution annual NTL products (high-definition VIIRS, VIIRS-HD) by synergistically integrating road density, NDVI, and water body data with raw VIIRS imagery. Evaluated across eight major regions in China (encompassing 341 cities), the VIIRS-HD product effectively mitigates the blooming effect inherent in raw VIIRS data, enabling the clear distinction and spatial variability of urban centers, suburban hubs, and undeveloped zones. Comparative analysis against three benchmark vegetation-adjusted NTL index downscaling methods and Luojia1-01 NTL data demonstrates the superior performance of our approach. At the city scale, VIIRS-HD maintains high fidelity to raw VIIRS NTL data (R 2 : 0.892–0.995). Pixel-level validation against Luojia1-01 NTL imagery across 14 cities shows that our product significantly outperforms raw VIIRS NTL, with the average R 2 increasing from 0.79 to 0.81. Furthermore, we produced a multi-temporal dataset (2014–2020) for China’s North Coast Region, confirming the method's robustness and scalability for long-term, large-area applications. This generated NTL product provides a robust dataset for monitoring urbanization processes and unveiling spatial details.